As a popular acquiring means of environment scenes, image collecting apparatuses are applied widely. The image collecting apparatuses may be applied aspects of various fields, such as video chatting, image recognition, teleconferencing, and real time monitoring etc.
The existing image collecting apparatuses collect original image information by an image sensor, and then process images by an image signal processing unit. The purpose of this processing is to obtain an image in which the human eyes are interested. For example, the human eyes like images having variety of color, thus the image signal processing unit needs a great deal of complex operations and modification to process original images, thereby obtaining images having variety of colors. For further example, the human eyes like a portrait having mellow facial features, thus the image signal processing unit will tone down the facial features of the portrait by employing some operations.
Furthermore, various image recognition algorithms may be also use the image processed by the image signal processing unit, to recognize and judge feature points, thereby obtaining information required by a machine. However, based on different application scenes, the information recognized by the image recognition algorithms may be different, for example, obtained landmark building information, text information etc.
In a procedure implementing technical solutions of embodiments of present disclosure, it finds there are following disadvantages in the prior art:
(1) In the prior art, since the original images collected by the image collecting apparatus need to be processed by the image signal processing unit as the images which the human eyes are interested, but the images which the human eyes are interested may generally give prominence to color information of the images but blur profile information, there is a technical problem in which the images collected by the image collecting apparatus lacks essential image information in some applications.
For example, in the image recognition algorithm, exaggerative lines and profiles are needed to maintain recognition accuracy. Since the images processed by the image signal processing unit may typically blur lines and profiles, this results in being lack of profiles and lines information, thus cannot satisfy the need of the image recognition algorithm.
(2) In the prior art, since the original images collected by the image collecting apparatus need to be processed by the image signal processing unit as the images which the human eyes are interested, but such images may lost a great deal of profiles and lines, this results in a lower accuracy for the image recognition algorithm.